藏北高原土壤湿度MODIS遥感监测研究

利用Terra与Aqua两颗卫星的MODIS地表温度和植被指数数据,分别构建LST-NDVI与LST-EVI共四种不同组合的Ts-VI特征空间,并依据该特征空间提取温度植被干旱指数(TVDI)作为反映土壤干湿状况的指标,探讨一种合适的遥感监测藏北高原土壤湿度的方案,并基于同步的实地土壤表层含水量采样数据进行验证评价。研究表明,四种数据组合方案提取的TVDI分布图均能较好反映藏北土壤湿度,且Terra MODIS LST-EVI构建的特征空间提取TVDI指标效果最佳。在此基础上获得的藏北高原土壤湿度分级图表明,从东南到西北土壤湿度逐渐降低,并呈现明显的空间分异规律。

The Study of MODIS Remote Sensing for Monitoring Soil Moisture over Northern Tibetan Plateau

The datasets of land surface temperature(LST) and vegetation index(NDVI and EVI) from Terra/Aqua MODIS products are used to construct four sorts of Ts-VI spatial patterns by different combinations of LST and VI,from which the Temperature-Vegetation Drought Index(TVDI) is extracted.The study attempts to develop a remote sensing monitoring method to indicate the levels of soil moisture over northern Tibet plateau by the TVDI.Then the evaluation of TVDI’s validity is verified by field sampling points of the studied area.The study shows that all the TVDI maps being extracted from four Ts-VI spatial patterns have an excellent reflection of local soil moisture status,and an obviously negative linear correlation between TVDI and sampling data is approved.The results also show the spatial pattern of Terra MODIS LST-EVI has a better performance to calculate TVDI than other combinations.Based on the TVDI,a classification map of soil moisture of northern Tibet plateau is made.From this map,it can be inferred that the level of soil moisture declines gradually from southeast to northwest of the studied area and apparent spatial heterogeneity exists.